Etiology of Genital Ulcers and Prevalence of Human Immunodeficiency Virus Coinfection in 10 US Cities
- Kristen J. Mertz,
- David Trees,
- William C. Levine,
- Joel S. Lewis,
- Billy Litchfield,
- Kevin S. Pettus,
- Stephen A. Morse,
- Michael E. St. Louis,
- Judith B. Weiss,
- Jane Schwebke,
- James Dickes,
- Romina Kee,
- James Reynolds,
- Don Hutcheson,
- Donata Green1,
- Irene Dyer,
- Gary A. Richwald,
- James Novotny1,
- Isaac Weisfuse,
- Martin Goldberg,
- Judith A. O'Donnell,
- Richard Knaup and
- Genital Ulcer Disease Surveillance Group
- Centers for Disease Control and Prevention, Atlanta, Georgia; Roche Molecular Systems, Alameda, California; University of Alabama at Birmingham; Chicago Department of Public Health, Illinois; Cincinnati Health Department, Ohio; Dallas County Health and Human Services and City of Houston Health and Human Services, Texas; Los Angeles County Department of Health Services, California; Memphis and Shelby County Health Department, Tennessee; New York City Department of Health, New York; Philadelphia Department of Public Health and Department of Medicine, Allegheny University of the Health Sciences, Philadelphia, Pennsylvania; St. Louis Department of Health and Hospitals, Missouri
- Reprints or correspondence: Dr. Kristen Mertz, Centers for Disease Control and Prevention, MS E-02, 1600 Clifton Rd., N.E., Atlanta, GA 30333.
-
Presented in part: 97th general meeting of the American Society for Microbiology, Miami Beach, May 1997 (abstract C-318).
Abstract
To determine the etiology of genital ulcers and to assess the prevalence of human immunodeficiency virus (HIV) infection in ulcer patients in 10 US cities, ulcer and serum specimens were collected from ∼50 ulcer patients at a sexually transmitted disease clinic in each city. Ulcer specimens were tested using a multiplex polymerase chain reaction assay to detect Haemophilus ducreyi, Treponema pallidum, and herpes simplex virus (HSV); sera were tested for antibody to HIV. H. ducreyi was detected in ulcer specimens from patients in Memphis (20% of specimens) and Chicago (12%). T. pallidum was detected in ulcer specimens from every city except Los Angeles (median, 9% of specimens; range, 0%–46%). HSV was detected in ≥50% of specimens from all cities except Memphis (42%). HIV seroprevalence in ulcer patients was 6% (range by city, 0%–18%). These data suggest that chancroid is prevalent in some US cities and that persons with genital ulcers should be a focus of HIV prevention activities.
Few data are available on the prevalence of genital ulcer pathogens in most US communities; however, surveillance data indicate that genital herpes is by far the most common ulcerative sexually transmitted disease (STD), with syphilis second and chancroid a distant third. Herpes was responsible for an estimated 208,000 initial visits to physicians' offices in 1996 [1]. During that year, 3831 cases of primary syphilis were reported to the Centers for Disease Control and Prevention (CDC), compared with 386 cases of chancroid [1]. Chancroid, however, is probably vastly underreported in the United States [2]. It is difficult to diagnose on clinical grounds alone, and most medical providers do not have the capability to perform laboratory tests for chancroid [3]. Outbreaks of chancroid may not be recognized. In Jackson, Mississippi, where chancroid had not been reported or even widely suspected, an investigation in 1994 and 1995 showed that Haemophilus ducreyi was the most common etiologic agent detected in ulcer specimens from patients at the STD clinic [4].
Ulcerative STDs have been associated with increased acquisition and transmission of the human immunodeficiency virus (HIV). Persons with genital ulcers are at higher risk for acquiring HIV than persons without ulcers [5, 6]. HIV-infected persons with genital ulcers may transmit HIV more efficiently than patients without ulcers [6]. Thus, recognition and control of chancroid, syphilis, and herpes may be important for HIV prevention. In Mwanza, Tanzania, improved diagnosis and treatment of STDs, including ulcerative STDs, was associated with a significant reduction in HIV seroconversion [7].
Because of our concern about the possibility of unrecognized chancroid in US cities, we performed a study to determine the etiology of genital ulcers in 10 cities. We used a research prototype multiplex polymerase chain reaction (M-PCR) assay that can detect DNA sequences from H. ducreyi, Treponema pallidum, and herpes simplex virus (HSV) from a single swab specimen with a sensitivity of > 90% for all 3 organisms [8]. The second objective of the study, given the increased risk for HIV in persons with genital ulcers, was to estimate HIV prevalence in ulcer patients in these 10 cities.
Methods
The study was performed at an STD clinic in 10 of the 11 US cities with the highest numbers of reported primary syphilis cases in 1994. Cities with high case loads of syphilis were selected to increase the chances of finding chancroid, given that syphilis and chancroid are associated with some of the same risk factors, including commercial sex and crack cocaine use [4, 9]. The start date of the study varied by clinic, from 4 January 1996 in Birmingham, Alabama, to 1 April 1996 in New York City.
At each of the 10 participating STD clinics, all patients with genital ulcers were asked to participate in the study, until 50 were enrolled. Patients with open sores were eligible; those with only unbroken vesicles were not. The base of the largest ulcer was swabbed for M-PCR testing. At some clinics, an extra tube of blood was drawn for HIV serologic testing at the CDC; at other clinics, discarded blood was used for this purpose. For each enrolled patient, a clinician recorded sex, age group (15–19, 20–24, 25–29, 30–34, 35–39, or ≥40 years), on-site rapid plasma reagin (RPR) and darkfield microscopy results, clinical diagnosis, and treatment. No identifying information was recorded on the data sheet or on the specimens. For persons who did not wish to participate, information on sex, age group, and on-site RPR result was recorded. All patients with ulcers received routine testing and treatment according to the usual clinic protocol, regardless of study participation.
The swab collected for M-PCR was swirled in a tube containing 1 mL of transport medium (Amplicor; Roche Diagnostic Systems, Branchburg, NJ) and then discarded. The tubes were stored at −20°C, transported frozen to CDC, and thawed before testing. The specimens were aliquoted, diluted, and tested in duplicate as described previously [8], except that 10 U of Taq polymerase (AmpliTaq; Perkin-Elmer Cetus, Norwalk, CT) was used. Three pairs of biotinylated oligonucleotide primers [8] were used to amplify segments of the H. ducreyi 16S rRNA gene, the T. pallidum 47-kDa membrane immunogen gene, and the HSV gB gene. PCR inhibition was analyzed by use of an internal control plasmid. Specimens containing inhibitors were extracted with phenol-chloroform and reassayed.
Serologic testing for HIV-1 antibodies using an EIA (Genetic Systems, Redmond, WA) with confirmation by Western blot (Cambridge Biotech, Worcester, MA) was performed at the CDC.
Results
At each participating clinic, 50–55 persons with genital ulcers were enrolled. Of those enrolled, 68% were men and 32% were women; 69% were < 35 years old. The study period varied from ∼3 months in Birmingham, New York, and Philadelphia to ∼10 months in Chicago. The number of ulcer patients refusing to participate varied by site, from none in Los Angeles to 29 in Chicago (median, 11). Persons refusing to participate were similar to those who consented with regard to sex, age, and on-site RPR results.
H. ducreyi was detected by M-PCR in specimens from patients in Memphis, where 20% of ulcer patients were positive, and in Chicago, where 12% were positive (table 1). In Chicago, clinicians had suspected chancroid and had prescribed recommended antibiotic therapy for 5 of the 6 patients with H. ducreyi. In Memphis, diagnosis and treatment information was available for 9 of 10 patients with detectable H. ducreyi; none were diagnosed with chancroid, and only 2 were given an antibiotic regimen recommended for chancroid.
Table 1
Results of the multiplex polymerase chain reaction (M-PCR) assay, by city.
The most commonly detected pathogen in the ulcer specimens was HSV, which was present in ≥50% of specimens (including those with multiple pathogens detected) at all sites except Memphis (42%). The percentage of ulcers with T. pallidum (including ulcers with multiple pathogens detected) varied by site, from 0% in Los Angeles to 46% in Memphis (median, 9%). The percentage of ulcer specimens with no detectable ulcer pathogens varied from 10% in Memphis to 32% in St. Louis; overall, 22% of the specimens were negative for DNA of all 3 organisms.
A higher percentage of men than women were infected with H. ducreyi (4% vs. 1%) and T. pallidum (15% vs. 7%). Of women with ulcers, 70% had HSV alone; of men, 59% had HSV alone. About the same percentage of men and women with ulcers had no detectable pathogen (23% and 22%, respectively). The prevalence of each organism varied by age group: 60% of H. ducreyi infections and 54% of T. pallidum infections occurred in persons ≥35 years of age. Most infections with HSV alone (76%) were detected in persons < 35 years old.
For patients receiving darkfield examinations, only 49% of persons positive for T. pallidum by M-PCR were positive by darkfield microscopy. At sites where RPR tests were done on-site during the patient's visit, 79% of patients positive for T. pallidum by M-PCR had reactive RPR results. Of 66 patients with T. pallidum detected by M-PCR, treatment information was available for 63; of these, 12 (19%) were not diagnosed with syphilis and were not given a recommended antibiotic regimen for syphilis. Of 75 patients with reactive RPR tests on site, 27 (36%) did not test positive for T. pallidum by M-PCR (14 had HSV only, and 13 had no pathogen detected).
Of the 516 persons with genital ulcers tested by M-PCR, 510 (99%) were tested for HIV. Of these, 31 (6%) were positive for HIV. HIV positivity varied by city, from 0% in Cincinnati to 18% in Philadelphia (table 2). HIV infection was most common among persons with only HSV detectable by M-PCR (22 [7%] of 316); 0 of 16 persons with H. ducreyi tested positive for HIV, 2 (4%) of 51 persons with T. pallidum only tested positive, and 0 of 12 coinfected with HSV and T. pallidum were HIV-positive. Of 114 persons with no organism detected, 7 (6%) were HIV-positive.
Table 2
No. and percentage of patients with genital ulcers who were HIV-seropositive, by city.
Discussion
Genital ulcers are difficult to diagnose on clinical grounds, and laboratory tests, especially for chancroid, are often not available [3]. In the United States, in areas where chancroid is endemic, clinicians may diagnose and treat it despite lack of routine testing; however, when chancroid spreads to other areas, it may be misdiagnosed until therapy fails in several ulcer patients [10]. In this study, chancroid was found in 2 cities, Chicago and Memphis; in these cities, H. ducreyi was detected in a significant percentage of ulcer specimens (12% and 20%, respectively). In 1 of these cities, clinicians did not suspect chancroid and therefore did not prescribe a recommended antibiotic for most infected persons.
The most widely available laboratory tests for STD clinic patients with genital ulcers are darkfield microscopy and serologic tests for syphilis. In this study, darkfield microscopy detected T. pallidum in 49% of the ulcers positive by M-PCR, a lower percentage than in a previous study [8]. Darkfield microscopy was not a focus of our study, but its low sensitivity in routine practice is noteworthy. Clinicians in many cities may lack darkfield experience because a small number of ulcer patients, primary syphilis patients in particular, attend the clinics.
The sensitivity of the RPR test compared with M-PCR was greater (79%) than that of darkfield microscopy, but still not close to 100%. Thus, several patients with a positive M-PCR result for T. pallidum were not treated for syphilis. In addition, the RPR test is not specific for primary syphilis; in this study, over one-third of patients with reactive RPR tests did not have detectable T. pallidum by M-PCR but probably had previously treated or latent syphilitic infection. Syphilis serologic tests are particularly nonspecific for infectious syphilis in an area of high syphilis morbidity [11].
No pathogen was detected by M-PCR in 22% of ulcer specimens; similarly, in a study in Jackson, Mississippi, 20% of ulcer specimens were negative by M-PCR [4]. These negative ulcers may have been caused by other recognized ulcerative diseases, such as lymphogranuloma venereum or granuloma inguinale, by trauma or superinfection, or by as yet unrecognized causes of genital ulcers. Alternatively, lack of serous fluid in the ulcer, inadequate specimen collection, improper storage, or delay in transport may have occurred, although we have no evidence of such problems.
When routine testing for the major ulcer pathogens is unavailable, it is important for clinicians to know what ulcer-causing organisms are present in their area so that they can make informed decisions regarding presumptive treatment. For example, if H. ducreyi is a significant cause of genital ulcers in a community, the CDC recommends presumptive therapy for both syphilis and chancroid when the diagnosis is unclear [12]. A periodic assessment of the prevalence of ulcer-causing pathogens may be useful in establishing or modifying local treatment protocols. Appropriate treatment regimens will facilitate healing and may reduce the risk for HIV infection or transmission.
Studies in US STD clinics show that HIV prevalence is higher in persons with genital ulcers than in those without ulcers [4, 13]. We found a high prevalence of HIV infection (6%) in the 510 ulcer patients tested for HIV in our study. Four clinics in our study (Chicago, Houston, Los Angeles, and New York) also conducted blinded HIV seroprevalence studies in 1996; at these clinics, HIV prevalence for the ulcer study patients was higher than for the general STD clinic population (1.2%, 1.9%, 1.8%, and 6.4%, respectively; CDC, unpublished data).
In the HIV-seropositive patients in our study, HIV acquisition probably preceded the development of the ulcers present at the time of the study. Patients may have had previous episodes of ulcers that facilitated transmission, or they may have engaged in high-risk sexual practices that exposed them over time to both HIV and ulcerative diseases. They may also have become infected with HIV from intravenous drug use.
HIV-positive persons with HSV may have had reactivated HSV, which is common in HIV-infected patients. In addition, HIV-infected persons may seek medical attention for HSV more frequently than HIV-negative persons because acyclovir-resistant strains and large persistent ulcerations are more common in immunocompromised persons [14]. Thus, it is not surprising that we found a high prevalence of HIV infection in ulcer patients with HSV DNA detected in genital lesions.
In view of the high prevalence of HIV in persons with genital ulcers in our study and the already established role of genital ulcers as a risk factor for HIV transmission, persons with ulcers must be treated quickly and appropriately to prevent transmission of both ulcerative STDs and HIV. Periodic surveillance of the etiology of genital ulcers in communities should enable providers to formulate appropriate empiric treatment regimens for ulcers. In addition to STD control, HIV counseling and testing for all genital ulcer patients is recommended to identify persons with HIV and to reduce HIV transmission.
Acknowledgments
We thank the staff of the 10 participating STD clinics for data and specimen collection. We also gratefully acknowledge Joe Courtney for beginning the study in Los Angeles and Janin Rockowitz and Sudha Mehta for assisting with the study in Philadelphia and Cincinnati, respectively. We thank Stuart Berman for proposing the study design, Orlando Blancato for facilitating communication with study sites, Karina Orle for preparing reagents, Martha Fears for performing syphilis serologic testing, and Charles Schable and his staff for performing the HIV tests.
Footnotes
-
Informed consent was obtained from all study participants. The study protocol was approved by the CDC Institutional Review Board and by local review boards.
- Received March 23, 1998.
- Revision received August 3, 1998.
- © 1998 by the Infectious Diseases Society of America
References
This Article
- AbstractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
Classifications
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 6.288
5-Yr impact factor: 6.103
For The Media
For Authors
- Services for authors
- Self-archiving policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
